Multiple spectral indices to predict the variability of structural and functional attributes in arid areas

Abstract

In the Argentinean Patagonia, the main driving forces of degradation are the extractive activities (oil and mining) and sheep ranching. Determine the ability of spectral indices to predict the spatial variability of structural and functional attributes of arid ecosystems would help to identify patterns associated with desertification or passive restoration. One of the main problems in these environments is their large territorial extension. Thus, remote sensing indices could be useful tools for assessing these areas. The aim of this study was to evaluate the ability of multiple spectral indices SPOT 5 satellite to predict the spatial variability of a structural attribute (perennial vegetation cover) and of a functional attribute (forage production) in arid areas. We assessed the relationship of seven spectral indices obtained from the satellite SPOT 5 in 39 sites located in three landscape units: coastal valleys (n=7), plateaus (n=18) and eastern valleys (n=18) in central Patagonia (Argentina). The green normalized difference vegetation index (GNDVI), the two modified soil adjusted vegetation index (MSAVI2), the ratio vegetation index (RVI) and the normalized difference vegetation index (NDVI) were the best predictors of perennial vegetation cover at landscape level. The spectral indices accounted for more than 34% of the variation in forage production in whole study area and for more than 60% in the western valleys. Our results show that spectral indices from satellite SPOT 5, mainly the GNDVI, the MSAVI2, the NDVI and the RVI, are appropriate tools to predict changes of the structure and function of the vegetation at landscape units level in arid zones.